Solve an ancient grammatical puzzle that has baffled scientists for 2,500 years
Solve an ancient grammatical puzzle that has baffled scientists for 2,500 years 11664
Researchers have been able to solve the grammatical puzzle that has defeated scholars since the fifth century BC.
Dr. Rishi Rajpupat, an Indian PhD student at the University of Cambridge, deciphered a rule devised by Panini, the “father of linguistics.” The rule is an essential part of an innovative grammar system created by Panini, dubbed the Language Machine, aimed at teaching India's sacred language Sanskrit.
Dr. Rajpuppat's efforts, detailed in his doctoral dissertation published today, now mean that the Panini language machine can be taught to computers for the first time.
Panini was a respected philologist and grammarian of ancient India, who lived sometime between the sixth and fourth centuries BC.
His "Language Machine" is widely considered one of the greatest intellectual achievements in history.
It is detailed in his esteemed work Aṣṭādhyāy, which is believed to have been written around 500 BC.
"It's one document and all it has is 4,000 very short rules," Dr Rajpuphat told MailOnline.
Each rule consists of three to four words on average. And what these 4,000 grammars do is they help us derive any word from the Sanskrit language.
These 4,000 rules basically work together like a machine.

Dr. Rajpuphat refers to it as a conceptual machine rather than a physical machine. The purpose of a language machine is to "derive" — form a word by changing the shape of the base or adding affixes to it.
When combining a base word with an suffix, there are differences in sound that need to be accounted for, otherwise they would give an illogical word like "definition" (pronounced def-ine-ey-shun).
The 4,000 rules that make up the Panini system effectively help users produce grammatically correct word forms.
Each rule has a serial number, based on its order in the document—for example, 7.3.103.
And in the event that a user discovers that two of these rules are applicable—a situation known as a "rule conflict"—Pāṇini creates one "metarule" to help users decide which of the two rules to apply.
Dr. Rajpuphat calls Panini's rule "1.4.2 vipratiṣedhe paraṁ kāryam".
And yet, the meaning of this "metarule" has been widely misinterpreted for about 2,500 years, leading to grammatically incorrect results.
Dr Rajpuphat said: “Unfortunately, the first researcher to comment on Panini's rules, Katyayana, misrepresented this rule. He was aware of two possible interpretations of this rule and unfortunately chose the wrong one. After that, all the scholars who have written about the Panini rules over the past 2,500 years have gone forward with this incorrect interpretation.”
The traditional but incorrect interpretation of rule 1.4.2 is that when two rules conflict, the rule with the higher sequential order or "win" should be chosen (eg, 7.3.103 instead of 7.1.9).
Dr Rajpuphat said: “Now this of course gives us kind of grammatically incorrect forms if we want to follow this interpretation. And I reinterpreted this rule as if such an interaction occurs between two rules in the same step, the rule that applies to the right part of the word wins. This helped us figure out the algorithm that runs this device, so now when you follow the correct interpretation of this rule, you automatically get the correct answer.”
And over the past 2,500 years, scientists have diligently developed hundreds of other metarules to try to fix the system and make it work — even though the system wasn't broken.

Dr Rajpupat said: “They had to come up with all sorts of extra instructions to help the grammarians come up with the grammatically correct form. Panini had an extraordinary mind, and he built a machine unparalleled in human history. He didn't expect us to add new ideas to his rules. The more we fiddle with the rules of Panini, the more we elude it.”
Dr Rajpuphat's work means we have a "very elegant, simple and teachable" algorithm that runs Panini's grammar that can be taught to computers.
Professor Vincenzo Vergiani, Dr Rajpuphat's thesis supervisor at Cambridge, said: "This discovery will revolutionize the study of Sanskrit at a time of increasing interest in the language."



Source : websites